For Peer Review - CORE · For Peer Review 3 our preceding publication (see the previous paper in this series). FTIR measurements were carried out with a Shimadzu FTIR-8300 equipped

For Peer Review

Rubbery Wound Closure Adhesives. II. Initiators for and

Initiation of 2-Octyl Cyanoacrylate Polymerization

Journal: Journal of Polymer Science, Part A

Manuscript ID: Draft

Wiley - Manuscript type: Original Article

Keywords: networks < N, rubber < R, adhesives < A, polyisobutylene, cyanoacrylate, initiators < I

ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901

Journal of Polymer Science, Part A: Polymer Chemistry

For Peer Review

1

Rubbery Wound Closure Adhesives. II. Initiators for and Initiation of 2-Octyl

Cyanoacrylate Polymerization

Istvan Szanka1,2

, Amalia Szanka1,2

, and Joseph P. Kennedy1

1 Institute of Polymer Science, The University of Akron, Akron, OH 44325-3909, USA 2 Polymer Chemistry Group, Institute of Materials and Environmental Chemistry, Research Centre for

Natural Sciences, Hungarian Academy of Sciences, H-1117 Budapest, Magyar tudósok krt. 2, Hungary

Correspondence to: Joseph P. Kennedy (E-mail: [email protected])

INTRODUCTION

Alkyl cyanoacrylates (CAs), typically butyl- or 2-

octyl cyanoacrylate (Bu-CA, OctCA), are

commercially successful wound closure

adhesives, marketed under a variety of trade

names, such as Hystoacryl®, Dermabond®, etc.

The active ingredient in gold standard

contemporary wound closure adhesives is

OctCA.

These CAs, due to the strongly electron-

withdrawing 2-cyanide and -ester substituents

at the vinyl function, are extremely reactive

monomers that rapidly polymerize anionically

upon contact with a variety of initiators,

including even the weakest nucleophiles.1-3

The first part of this series of publications

concerned the design, synthesis,

characterization, structure, and in vitro testing

of rubbery homo- and conetworks prepared

with Ø(PIB-CA)3, and OctCA plus Ø(PIB-CA)3,

respectively, as potential occlusive wound

closure adhesives (see the first paper of this

series of publications). We analyzed the

advantages/disadvantages of known alkyl-CA

skin adhesives, and demonstrated that

flexibilizing them by the covalent incorporation

of rubbery PIB moieties greatly improves their

applicability as potential wound closure

adhesives. We continued research in this area

and determined the “stir-stop time” and “set

time” of select CA polymerization initiators. We

define stir-stop time the time the viscosity of a

liquid wound closure adhesive increases to the

point where stirring stops, and set time the time

a liquid wound closure adhesive becomes an

essentially tack free skin closure. Set times are

ABSTRACT

The polymerization of 2-octyl cyanoacrylate (OctCA) initiated by five N-bases [N,N-dimethyl-p-

toluidine (DMT), pyridine (Pyr), triethyl amine (Et3N), azobicyclo[2.2.2]octane (ABCO), and

diazobicylo[2.2.2]octane (DABCO)] was investigated. Our main objective was to assess the suitability

and relative reactivity of these initiators for neat OctCA polymerization as wound closure adhesives.

Methodologies were developed to determine stir-stop and set times of OctCA polymerization and to

use these quantities to assess initiation reactivity. According to these studies Et3N, ABCO, DABCO,

and Pyr are most reactive initiators, while DMT is much less reactive. Polymerizations were much

faster in the presence of small amounts of tetrahydrofuran than toluene, indicating solvent polarity

effects. Initiator reactivity is discussed in terms of structural parameters. NMR and MALDI-TOF

analyses of low molecular weight P(OctCA) prepared with DMT did not show evidence for the

expected aromatic head group proposed by earlier investigators, which suggests complex initiation

mechanism.

KEYWORDS: network, rubber, adhesives, polyisobutylene, cyanoacrylate, initiators

Page 1 of 19



For Peer Review

2

of great practical significance for health

professionals who use liquid wound closure

adhesives. Surgeons and nurses prefer set times

in the 30-120 seconds range.

Wound closure adhesives are packaged in

special delivery devices in which the

cyanoacrylate (plus a variety of additives) is

sealed in a thin glass capsule that is crushed

upon deployment, forcing the liquid monomer

through a small porous sponge (typically

polypropylene or nylon) at the delivery port

onto the skin. Figure 1 shows such a device.

FIGURE 1 A typical delivery device for

cyanoacrylate wound closure adhesives. The

active ingredients (plus various additives, i.e.,

stabilizer, viscosifying agent, dye, etc.) are in a

thin walled glass ampule, which, when broken,

sends the liquid through the sponge onto the

skin.

It is not generally known that the sponge at the

tip of the device performs two critical functions:

(a) it contains an initiator, a key ingredient that

induces the polymerization of the CA as it is

squeezed through the sponge, and (b) it helps

delivering the adhesive evenly over the

targeted surface. Absent the initiator, set time

is significantly lengthened, which may render

the adhesive practically useless. The set time of

a commercially acquired Dermabond® sample

was 50 ±10 seconds after the glass capsule was

shattered and the liquid squeezed through the

sponge onto a glass surface. In contrast, the set

time increased to 8-10 minutes when the vial

was shattered and the liquid spilled directly

over the same surface without contacting the

sponge. A Dermabond® sample remained liquid

for over two weeks after being exposed to

humid air in an open vial.

The nature and concentration of initiators used

for the polymerization of OctCA in commercial

devices are closely guarded secrets and our

search of the scientific literature to identify

specific initiators remained fruitless. Searching

the patent literature gave some clues as to the

identity of initiators, and other ingredients

(plasticizing agents, stabilizers, thickeners, pH

modifiers, preservatives, dyes, etc.) used in

combination with OctCA in commercial

formulations.

Herein we describe investigations to gain insight

into the mechanism of initiation of CA

polymerizations by Et3N, ABCO, DABCO, Pyr,

and DMT, and to assess their relative

polymerization reactivities. Our findings are of

interest for controlling the set time of OctCA,

and for the homopolymerizations of Ø(PIB-CA)3

and copolymerizations of OctCA plus Ø(PIB-CA)3

to flexible wound closures (see the previous

paper in this series). This paper also concerns an

assessment of the relative reactivities of the

above five bases for OctCA polymerization,

efforts to elucidate the mechanism of initiation

of OctCA polymerization, and a discussion of

the results in terms of structural considerations.

EXPERIMENTAL

Materials and Procedures

Azobicyclo[2.2.2]octane (ABCO, Alfa Aesar),

diazobicyclo[2.2.2]octane (DABCO, Alfa Aesar),

triethyl amine (Et3N, Aldrich), and pyridine (Pyr,

Fischer Chem.) were used as received without

further purification. Additional materials,

instrumentation, and procedures have been

described in the first paper of this series (see

the previous paper in this series).

GPC and 1H NMR instrumentation and

associated procedures have been described in

Page 2 of 19



For Peer Review

3

our preceding publication (see the previous

paper in this series).

FTIR measurements were carried out with a

Shimadzu FTIR-8300 equipped PIKE MIRacle

Attenuated Total Reflectance (ATR) accessory

with diamond crystal plate. A drop of liquid

adhesive was placed on the diamond crystal,

and IR measurements were acquired at room

temperature. Thirty scans were obtained and

averaged with 8 cm-1 resolution.

MALDI-TOF spectra were recorded on a Bruker

Reflex II time-of-flight mass spectrometer

equipped with a nitrogen laser (Laser Science)

operating at 337 nm and 20 Hz repetition rate.

To facilitate mass analysis, efforts were made to

reduce the molecular weight of P(OctCA)

samples. Thus, OctCA polymerizations were

carried out at low monomer concentrations

(e.g., [M]/[I] = 10), however, the Mn could not

be decreased below ~20,000 g/mol. The MALDI

matrix solution was prepared by dissolving 2-

[(2E)-3-(4-tert-butylphenyl)-2-methylprop-2-

enylidene]malononitrile (DCTB) and potassium

trifluoroacetate (100:10) in dry N,N-

dimethylformamide (DMF) (20 mg/mL).

Samples were prepared by depositing 0.5 mL

matrix solution on the stainless steel MALDI

target, drying it in a gentle stream of air, placing

on top of the dried matrix a layer of 0.5 mL

P(OctCA) solution (20 mg polymer/mL

dichloromethane), and drying the system in a

gentle stream of air. For the sandwich method a

second matrix layer (using 0.5 mL matrix

solution) was deposited on top of the dried

polymer layer, producing a matrix-sample-

matrix sandwich.

Stir-Stop Time

One way to assess the reactivity of initiators is

in terms of stir-stop time, i.e., the time (in

seconds) liquid OctCA becomes a non-stirrable

mass upon the addition of a well-defined

amount (moles) of initiator at room

temperature. Thus, experiments were carried

out by placing in a 5 mL flat-bottom screw cap

vial 1 mL OctCA and a Teflon coated small (1

cm) magnetic stir bar. At time = 0 a known

amount of initiator (in certain cases dissolved in

toluene or THF) was added to the monomer by

a microsyringe, the vial was quickly capped,

manually strongly agitated for 1-2 seconds,

placed on a stirring plate, and stirred at ~60

rpm. Stir-stop time (in seconds) was recorded

when stirring suddenly stopped (the charge

“froze”) due to the increased viscosity of the

system. (See also the next section for a method

to determine monomer conversion at gel time.)

Polymerizations induced by all five initiators

were highly exothermic and experiments were

carried out to follow heat evolution. In these

experiments the reactor vial was hermetically

capped with a rubber septum through which a

thermocouple was inserted into the charge, and

the heat evolved after initiator addition was

followed as a function of time. Figure 2 shows

the temperature-time profile of a

representative experiment. The process starts

with a lengthy induction period (may be several

hundred seconds), in the course of which the

initiator consumes the SO2 stabilizer in the

monomer. Subsequently, rapid exothermic

polymerization ensues. During the

polymerization the viscosity of the charge

increases very rapidly until a mass of gooey

polymer forms and stirring becomes impossible.

The stir-stop time, i.e., the time when stirring

stops, is viewed as a measure of initiator

reactivity.

FIGURE 2 Temperature-time profile of a

representative neat OctCA polymerization

Page 3 of 19



For Peer Review

4

(initiator: Pyr dissolved in toluene,

[M]/[I]=8400)

Set Time

A generally accepted definition of set time is

lacking. We developed a test for the

determination of set times that mimics practical

clinical requirements. The test consists of

placing 1 mL OctCA in a 5 mL flat-bottom vial,

injecting an appropriate amount of initiator by a

microsyringe, manually vigorously agitating the

system for 1-2 seconds, and immediately

coating the liquid on a ~2 x 10 cm strip of fresh

ventral porcine skin. After visually examining

the coat, the surface of the coat is lightly

touched with a latex-gloved index finger. Set

time (in seconds) is recorded when the

sensation felt touching the surface of the

coating is similar to that touching the surface of

a PostIt Self-stick removable paper. This test

gives reproducible time readings even when

practiced by less experienced operators.

Experiments were carried out to determine

monomer conversions at stir-stop and set

times. Thus, to 1 g OctCA stirred in a 5 mL flat-

bottom vial containing a small stir bar was

injected a desired quantity of DMT initiator and

the system was stirred. During the induction

period a small drop (~0.05 g) of the liquid was

removed from the charge by a Pasteur pipet

and placed on top of the probe in the Teflon

ring of the FTIR instrument. A FTIR spectrum

was obtained and the absorption at 3124 cm-1

associated with =CH2 stretch was analyzed

(Spectrum A, Figure 3). Stirring was continued

and a second spectrum was acquired at stir-

stop time (Spectrum B, Figure 3). The lag

between stir-stop time and acquiring the

spectrum was ~5 seconds. According to the

spectrum (and taking the sensitivity of the

instrument into consideration), monomer

conversion at stir-stop time was ≥90%.

Subsequently, the surface of the product in the

FTIR instrument was probed by lightly touching

it with the tip of a needle to ascertain that the

gooey mass changed to a hard solid, i.e., set has

occurred.

FIGURE 3 FTIR spectra of a representative

charge during the induction period (A), and at

stir-stop time (B).

RESULTS AND DISCUSSION

OctCA Polymerizations with N,N-Dimethyl

Toluidine (DMT)

The scientific and patent literatures were

searched to identify highly reactive anionic

initiators for alkyl-CA polymerization in general

and OctCA in particular. Among the many kinds

of initiators (e.g., amines, pyridines,

phosphines, alkonium salts, etc.) initially we

employed DMT because of its demonstrated

reactivity and initiating efficiency, ease of

handling, miscibility with CAs, and low price.4-8

Subsequent research substantiated

expectations in respect to DMT, and our earlier

homo- and copolymerization research was

carried out with this initiator (see the previous

paper in this series). In view of the considerable

experience gained with DMT, this base became

our “control” searching for more reactive

initiators.

There is agreement in the literature in regard to

the mechanism of initiation of CA

polymerization induced by DMT. According to

numerous investigators9-11 DMT initiates the

polymerization of CAs by Michael addition to

the monomer, followed by anionic chain growth

(addition) polymerization by the zwitterion:

Page 4 of 19



For Peer Review

5

CH2 C

CN

OO

CH

H3C (CH2)5CH3

C

OO

CH

H3C (CH2)5CH3

+

H2C CN

H3C N

CH3

CH3

H3C N

CH3

CH3

Support for this mechanism was the finding that

poly(4-vinylpyridine) initiates the graft

polymerization of Et-CA.11 Further, NMR and UV

spectroscopic investigations also substantiated

initiation by zwitterions.11-13 Pepper and

coworkers discussed the mechanism and

kinetics of Et- and Bu-CA polymerizations

initiated by various tertiary amines, and

speculated about the nature of the zwitterion

initiating species, and active chain ends (free

ions, tight ion pairs) involved in these

polymerizations.1,11-15

An obvious consequence of zwitterion initiation

with DMT is that the polycyanoacrylates formed

should contain an aromatic head group. In the

course of these investigations we analyzed by 1H NMR and MALDI-TOF spectroscopies the

structures of P(OctCA)s obtained under various

conditions with DMT, however, failed

unambiguously to identify the end groups.

1H NMR and MALDI-TOF Analysis

FIGURE 4 1H NMR spectrum of a representative

poly(OctCA) (prepared with DMT [M]0/[I]0=60)

Efforts were made to elucidate the mechanism

of initiation of OctCA polymerization induced by

DMT. If initiation of CA polymerization by N

bases occurs via zwitterions (see above), the

polymerization of OctCA by DMT should

produce the CH3-pC6H4-N+(CH3)2- head group ,

readily discernible by 1H NMR and MALDI-TOF

analysis. To facilitate spectroscopic analysis, we

prepared P(OctCA)s using relatively low DMT

concentrations (e.g., [M]/[I] = 10/1) to reduce

the molecular weight of the polymer.

Surprisingly, neither high resolution 1H NMR

spectroscopy nor MALDI-TOF analysis indicated

the presence of the expected aromatic head

group.

Thus, Figure 4, the 1H NMR spectrum of a

representative poly(OctCA), shows the absence

of protons in the 7.0-8.5 ppm (i.e., protons on

the heteroaromatic ring) range. The singlets at

6.59 and 7.02 ppm due to olefinic protons of

OctCA indicate the presence of small amounts

of unreacted monomer. The resonances at 4.27

and 5.00 ppm are due to the protons of 1-octyl

(-OCH2(CH2)6CH3) and 2-octyl (-

OCH(CH3)(CH2)5CH3) groups, respectively. The

resonance at 3.69 ppm suggests the presence of

HO-CH2- groups.

Further, Figure 5 shows the MALDI-TOF

spectrum of P(OctCA) acquired under standard

conditions (linear mode). Assuming singly

charged potassiated ions i.e., [polymer + K]+,

the resolved sharp peaks (indicated with mass

numbers) ranging from 3613 (n=17) to 8481 Da

(n>40) and beyond, are due to individual OctCA

oligomers. The maximum peak intensity at m/z

= 4451 corresponds to the 21-mer (calc. =

4450.95). Figure 6 shows a more detailed view

of the 21-mer and its satellites obtained in

reflector mode. These data can be reconciled by

assuming single potassiation leading to

[polymer-K]+ ions and initiation by HO- ions

(adventitious moisture):

Page 5 of 19



For Peer Review

6

C

OO

CH

H3C (CH2)5CH3

H2C CN

n

CH2 C

CN

OO

CHH3C

(CH2)5CH3

HO CH2 CH

CN

OO

CHH3C

(CH2)5CH3

CH2 C

CN

OO

CHH3C

(CH2)5CH3

HO CH2 C

CN

OO

CHH3C

(CH2)5CH3

+ H+

+ OH-

C

OO

CH

H3C (CH2)5CH3

CH2 CN

HO

nn

C

OO

CHH3C (CH2)5CH3

H2C CN

Mass spacing and absolute masses of the

resolved oligomers provide additional structural

information. Specifically, the mass spectrum

indicates at least two other oligomer series

having a mass spacing of 209.15 Da (the mass of

OctCA), and differing by ±30 Da from the major

component. The spectra do not show a single

mass distribution of homologous reaction

products and the nature of end groups is

unclear.

Evidently, these 1H NMR and MS analyses did

not provide the expected evidence of initiation

by zwitterions of OctCA polymerization. We are

compelled to conclude that initiation of OctCA

polymerization by DMT involves a series of

unidentified complex steps, and the absence of

aromatic head group indicates that the first

step of propagation does not occur by

zwitterions (see above). Some of the findings

may be interpreted to suggest initiation by HO-

(i.e., traces of moisture).

FIGURE 5 MALDI-TOF Spectrum of Poly(OctCA)

Prepared Under Standard Conditions (linear

mode).

FIGURE 6 Part of the Mass Spectrum of

P(OctCA). Besides the expected mass of 4451

Da two additional signals appear with a mass

difference of ±30 Da (= HCHO) (reflector mode).

Polymerization with various initiators

In addition to DMT, we investigated Pyr, Et3N,

ABCO, and DABCO as initiators for neat OctCA

polymerization:

N

N,N-Dimethyl-p-toluidine

(DMT)

N

Pyridine

N

Triethylamine

N

N

1,4-Diazabicyclo[2.2.2]octane

(DABCO)

N

1-Azabicyclo[2.2.2]octane

(ABCO)

Except for ABCO, these molecules, have been

used by earlier investigators and were found

highly reactive inducing the polymerization of

alkyl CAs. Among the tertiary aliphatic amines

ABCO appeared particularly promising because

of its high nucleophilicty and unencumbered N

atom. We thought, erroneously as it turned out

(see below), that ABCO will be more reactive

than Et3N because the three Et groups in the

latter may sterically hinder the approach of the

monomer during initiation. According to ab

initio calculations steric hindrance severely

retards polymerization of Et-CA (16). In contrast

to Et3N, such steric hindrance would be absent

with ABCO because the three -CH2-CH2- groups

on the N atom are forced out-of-the-way

Page 6 of 19



For Peer Review

7

rendering the N’s electron pair unencumbered

for initiation.

DABCO with two unencumbered N atoms was

selected for similar reasons, and because of its

low price. DABCO was used by a previous team

to initiate the polymerization of various alkyl

CAs16; however, these workers used aqueous

buffers as reaction media, which obscures their

results since water is an initiator of CA

polymerization.3

Since ABCO and DABCO are crystalline, they had

to be dissolved in an inert solvent (i.e., toluene

or THF) for using them as initiators. Dissolving

and using these solids in toluene or THF lead to

valuable observations in regard to solvent

polarity effects (see below).

We were also interested examining the

initiating reactivity of pyridine, as pyridine and

its derivatives were found to induce CA

polymerizations.2,14,15 However, we could not

carry out neat OctCA polymerizations with

pyridine because of the very high reactivity of

this base. The moment a smallest drop of

pyridine was added to OctCA, a solid blob of

polymer formed around the drop, and

homogeneous mixing became impossible. Thus,

pyridine had to be diluted with toluene or THF

to generate data.

Stir-Stop Time and Set Time

While stir-stop time and set time are

phenomenologically distinguishable (stoppage

of stirring at stir-stop time, surface dryness at

set time, see Experimental), their definition in

polymer chemical terms is challenging.

Repeated efforts have been made to determine

monomer conversion, P(OctCA) molecular

weights, and molecular weight distributions at

stir-stop and set times at various [M]0/[I]0

ratios, however, the differences between these

quantities were negligible (within experimental

error). For instance, we found by FTIR

spectroscopy that monomer conversions were

≥90% at stir-stop times, and that the spectra did

not change upon reaching set time. Similarly,

product molecular weights and distributions

obtained at stir-stop and set times were

indistinguishable.

The phenomenological difference between stir-

stop and set times may be due to the rate of

heat dissipation after polymerization cessation.

OctCA polymerization is extremely rapid and

highly exothermic (see Figure 2) and the

dissipation of the heat of polymerization to the

surrounding medium takes time. At stir-stop

time the temperature of the system is way

above ~53oC, the Tg of P(OctCA), and the

products are hot gooey masses due to the

accumulated heat of polymerization. After the

heat is dissipated, the gooey mass cools to

room temperature, and the product becomes

hard and dry to the touch. Termination is likely

due to protonation of the active growing ends

by adventitious moisture. A small amount of

chain growth may occur after stir-stop (i.e.,

after ≥90% conversion), however, this

incremental growth is negligible and cannot be

detected by GPC molecular weight

determination.

Figure 7 summarizes stir-stop times obtained

with various initiators. Contrary to convention

but to better visualize the findings, we plotted

[monomer]/[initiator] versus stir-stop time

(instead of plotting, as customary, the

dependent (stir-stop time) versus the

independent ([monomer]/[initiator]) variable).

Evidently, DMT is the least reactive initiator

among those examined, as it requires relatively

high initiator concentrations (low [M]0/[I]0

ratios) for polymerization. Increasing the

polarity of the system, i.e., changing from neat

monomer to a charge that also contains THF

introduced with DMT, affects initiation

efficiency relatively little. The inset in Figure 7

shows the effect of THF (i.e., 5 vol.% DMT

dissolved in THF), obscured by the larger

dimensions of the main plot. In contrast to

DMT, polymerizations induced by ABCO and

DABCO are quite sensitive even to slightly

increasing the polarity of the system i.e.,

introducing THF with the initiator.

Page 7 of 19



For Peer Review

8

FIGURE 7 Relative Reactivities (Stir-Stop Times)

of Various Initiators.

The fact that the straight lines in Figure 7 do not

extrapolate to the origin is likely due to the

presence of the ~ 30 ppm SO2 in OctCA (i.e., the

stabilizer added to ensure absence of

polymerization during storage). Polymerizations

are inhibited until a sufficient concentration of

anionic species arises to start the reactions.

The data obtained with ABCO and DABCO are of

interest because: (a) the initiating reactivities of

these tertiary cyclic aliphatic amines dissolved

in toluene and even more so in THF, are far

superior to that of DMT, (b) their reactivities are

very similar (within experimental error) both in

toluene and in THF, and (c) their reactivities are

far higher when dissolved in THF than in

toluene.

The intersections of the experimental plots with

the vertical dotted line at 300 sec provide a

means of comparison of the relative reactivities

of the initiating systems. By this measure the

order of initiating reactivities in the presence of

THF is: DMT << Pyr < Et3N ~ ABCO ~ DABCO,

while in the presence of toluene this order is

DMT << Et3N < ABCO ~ DABCO < Pyr. Evidently,

even a very small amount of polar solvent used

to dilute the initiator strongly affects relative

initiator reactivities.

Structural considerations explain these

observations: (a) the nucleophilicty of the

aliphatic cyclic amines, ABCO and DABCO, is

higher than that of DMT due to the electron

releasing effect of the aliphatic groups (i.e., the

aliphatic substituents in ABCO and DABCO are

more electron releasing than the aromatic

group in DMT); (b) only one of the two N atoms

in DABCO is probably active, rendering ABCO

and DABCO similar in reactivity. The likely

reason is that after one of DABCO’s N atoms

donates its electron pair and becomes positively

charged, the zwitterion is reluctant to donate its

second N’s electron pair since this would create

an energetically unfavorable second positive

site in the molecule.

CONCLUSIONS

Methodologies were developed to measure

initiator reactivity for neat OctCA

polymerizations induced by select N bases.

Polymerizations are very rapid and exothermic

upon contact with Et3N, ABCO and DABCO, and

Pyr, and much slower with DMT initiator.

Overall rates are faster with initiators dissolved

in THF than in toluene indicating polarity

effects. Relative initiator reactivity is discussed

in terms of structural parameters. According to 1H NMR and MMALDI-TOF spectroscopy

P(OctCA)s obtained by the use of DMT did not

incorporate the expected DMT head group,

which suggests complex initiation. Evidence for

initiation by adventitious moisture is presented

and discussed. It is of practical significance that

the set time of OctCA polymerization can be

precisely controlled in the 60-120 seconds

range by the use of appropriate concentration

of select initiators. Finally, let it be emphasized

that the polymerization of OctCA, i.e., the

formation of P(OctCA) wound closure coats, is

not “curing”, as it is often incorrectly referred

to, as the coat is not crosslinked (not a

network), but a soluble stiff thermoplastic.

ACKNOWLEDGEMENTS

We wish to acknowledge Ian McCullough’s

professional help searching the patent and

scientific literature, and Chris Wesdemiotis for

acquiring and interpreting MALDI-TOF spectra.

REFERENCES AND NOTES

Page 8 of 19



For Peer Review

9

1. E.F. Donnely, D.S. Johnston, D. C. Pepper, D.J.

Dunn, J. Polym. Sci. Pol. Lett. 1977, 15, 399-405.

2. J.P. Cronin, D. C. Pepper, Makromol. Chem.

1988, 189, 85-102

3. I.C. Eromosele, D.C. Pepper, B. Ryan,

Macromol. Chem. 1989, 190, 1613-1622.

4. C.J. Buck, (Johnson & Johnson), U.S. Patent.

3,903,055, Sept 2, 1975

5. J.P. Kennedy, Y. Kwon, S. Jewrajka U.S.

Patent. 2010/0144996, Jun 10, 2010

6. Y. Kwon, J. P. Kennedy, Polym. Advan.

Technol. 2007, 18, 800-807.

7. Y. Kwon, J. P. Kennedy, Polym. Advan.

Technol. 2007, 18, 808-813.

8. K. N. Broadley, J. Guthrie, N. Swords U. S.

Patent. 2010/0035997, Feb 11, 2010

9. M. G. Han, S. Kim, S. X. Liu, Polym. Degrad.

Stab. 2008, 93, 1243-1251.

10. D.S. Johnston, D.C. Pepper, Makronol Chem.

1981, 182, 407-420.

11. D.S. Johnston, D.C. Pepper, Makromol.

Chem. 1981, 182, 393-406.

12. D.S. Johnston, D.C. Pepper, Makromol.

Chem. 1981, 182, 421-435.

13. D.C. Pepper, Polym. J. 1980, 12, 629-637.

14. D. C. Pepper, R. Bernard, Makronolekul.

Chem. 1983, 184, 395-410.

15. C. Lochen, N. Otte, Macromolecules 2010,

43, 9674-9681.

16. D. Katti, N. Krishnamurti, J. Appl. Polym. Sci.

1999, 74, 336-344.

Page 9 of 19



For Peer Review

1

FIGURE 1 A typical delivery device for cyanoacrylate wound closure adhesives. The active ingredients

(plus various additives, i.e., stabilizer, viscosifying agent, dye, etc.) are in a thin walled glass ampule,

which, when broken, sends the liquid through the sponge onto the skin.

Glass Ampule

OctCA

(+other ingredients)

Sponge

Page 10 of 19



For Peer Review

2

FIGURE 2 Temperature-time profile of a representative neat OctCA polymerization (initiator: Pyr

dissolved in toluene, [M]/[I]=8400)

Page 11 of 19



For Peer Review

3

FIGURE 3 FTIR spectra of a representative charge during the induction period (A), and at stir-stop time

(B).

Page 12 of 19



For Peer Review

4

CH2 C

CN

OO

CH

H3C (CH2)5CH3

C

OO

CH

H3C (CH2)5CH3

+

H2C CN

H3C N

CH3

CH3

H3C N

CH3

CH3

Page 13 of 19



For Peer Review

5

FIGURE 4 1H NMR spectrum of a representative poly(OctCA) (prepared with DMT [M]0/[I]0=60)

Page 14 of 19



For Peer Review

6

C

OO

CH

H3C (CH2)5CH3

H2C CN

n

CH2 C

CN

OO

CHH3C

(CH2)5CH3

HO CH2 CH

CN

OO

CHH3C

(CH2)5CH3

CH2 C

CN

OO

CHH3C

(CH2)5CH3

HO CH2 C

CN

OO

CHH3C

(CH2)5CH3

+ H+

+ OH-

C

OO

CH

H3C (CH2)5CH3

CH2 CN

HO

nn

C

OO

CHH3C (CH2)5CH3

H2C CN

Page 15 of 19



For Peer Review

7

FIGURE 5 MALDI-TOF Spectrum of Poly(OctCA) Prepared Under Standard Conditions (linear mode).

Page 16 of 19



For Peer Review

8

FIGURE 6 Part of the Mass Spectrum of P(OctCA). Besides the expected mass of 4451 Da two additional

signals appear with a mass difference of ±30 Da (= HCHO) (reflector mode).

Page 17 of 19



For Peer Review

9

N

N,N-Dimethyl-p-toluidine

(DMT)

N

Pyridine

N

Triethylamine

N

N

1,4-Diazabicyclo[2.2.2]octane

(DABCO)

N

1-Azabicyclo[2.2.2]octane

(ABCO)

Page 18 of 19



For Peer Review

10

FIGURE 7 Relative Reactivities (Stir-Stop Times) of Various Initiators.

Page 19 of 19



Documents

For Peer Review - CORE · For Peer Review 3 our preceding publication (see the previous paper in this series). FTIR measurements were carried out with a Shimadzu FTIR-8300 equipped